Downloaded from http://sp.lyellcollection.

org/ at Carleton University Library on June 18, 2015

Origin and petrophysics of dolostone pore space

F. J E R R Y L U C I A
Bureau of Economic Geology, John A. and Katherine G. Jackson School of Geosciences,
The University of Texas at Austin, Austin, TX 78713-8924, USA

Abstract: The common claim that dolomitization creates 12% porosity is based on the
mole-for-mole replacement equation. However, in the past 50 years, data have been col-
lected demonstrating that dolomitization does not create porosity. Instead, dolostones
inherit the porosity and fabric of the precursor limestone, and porosity is reduced by over-
dolomitization. The porosity of the precursor limestone depends on the diagenetic history
up to the time of dolomitization. Data show that: (1) carbonates are born with high poros-
ity and lose porosity gradually over a long period; and (2) mud-dominated fabrics compact
more readily than grain-dominated fabrics. The problem of estimating the time of dolomi-
tization is minimized by confining observations to young limestones and associated dolo-
stones. Limited data from Holocene dolomitic sediments suggest no change in porosity with
dolomitization. Study of Plio-Pleistocene carbonates in Bonaire, Netherlands Antilles,
demonstrates that precursor limestones are more porous than dolostones. Limestones
average 25% porosity, whereas dolostones average 11% porosity. Data from the Neogene
of the Great Bahama Bank show that dolostones and adjacent limestones both have 40%
porosity. Porosity studies of the Jurassic Arab D Formation show that dolostones and
associated grain-dominated limestones have similar porosity ranges and that the decrease
in porosity with increasing dolomitization results from compaction of the mud-dominated
fabrics. These data suggest that porosity in dolostone is not created by a mole-for-mole
replacement mechanism, instead, dolostone porosity is: (1) inherited from the precursor
limestone; and (2) occluded by the process of overdolomitization. Palaeozoic dolostones,
however, are commonly more porous than juxtaposed limestones. The explanation for this
observation is that limestones lose porosity through compaction and cementation, whereas
dolostones resist compaction and retain much of their porosity.
Permeability studies have demonstrated that dolomitization of grain-dominated lime-
stones usually does not change porosity-permeability relationships. Instead, precursor
limestone fabric controls pore-size distribution. The dolomite crystal size of a mud-
dominated dolostone may, however, be larger than the carbonate mud size, improving the
porosity-permeability relationship substantially. Hence, there is a predictable relationship
between interparticle (grains or crystals) porosity, permeability, precursor grain size and
dolomite crystal size.

The origin and petrophysics of dolostone have
been linked since 1837, when Elie De Beaumont
proposed that the conversion of limestone to
dolostone creates 12% porosity because the
molar volume of dolomite is smaller than the
molar volume of calcite (De Beaumont 1837).
This chemical model was used to explain the
presence of vuggy pore space in dolostones of
the Tyrolean Alps, and after 165 years it is still
the one most commonly used to explain porous
dolostones. Although challenged by Landes
(1946), Schmoker & Halley (1982), Halley &
Schmoker (1983) and, most recently, by Lucia
& Major (1994), it is nonetheless the most
common dolomitization model taught in
colleges and universities (Tucker 2001).
The purpose of this paper is to demonstrate
that dolostone porosity is not the result of
dolomitization and that the concept that porous
dolostone is formed from non-porous limestone
is incorrect. Data will be presented to show that
many young limestones are very porous and
that this porosity is inherited by replacement
dolostone. Furthermore, the observation that
many Palaeozoic limestones are less porous
than associated dolostones will be shown to
result from differential compaction, limestones
being more susceptible to compaction than
dolostones. The excellent petrophysical quality
of many dolostones will be related to: (1) an
increase in crystal size in dolomitized mud-
dominated fabrics; and (2) the ability of dolo-
stone to retain porosity.
Terminology
Terminology is discussed here briefly because its
improper use adds to the confusion that

From: BRAITHWAITE,C. J. R., RIzzI, G. & DARKE, G. (eds) 2004. The Geometry and Petrogenesis of Dolomite
Hydrocarbon Reservoirs. Geological Society, London, Special Publications, 235, 141-155. 0305-8719/$15.00
9 The Geological Society of London 2004.
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
142 F.J. LUCIA
surrounds this issue. Porosity is defined as pore
volume divided by bulk volume and in
geological literature is commonly expressed as
per cent porosity, i.e. porosity • 100. Pore space
or pore volume is that portion of the bulk
volume not occupied by minerals. When the
term porosity is used, the bulk volume of the
sample should be clearly understood. In core
analysis, the bulk volume is the volume of the
sample measured. In thin section, the bulk
volume is the volume of the thin section.
However, in many instances, the bulk volume is
not clear and may not even be considered, as in
such comments as 'intergrain porosity' or 'a
description of porosity'. Intergrain or inter-
crystal porosity is the pore volume between
grains or crystals divided by the bulk volume
under consideration and is expressed as a
number. It cannot be described except by a
number or a reference to a value (high porosity,
for example). It is correct to refer to intergrain
porosity, such as in 'the intergrain porosity is
25%'. It is incorrect to use porosity with a
descriptor, such as 'large and well-connected
porosity', but it is correct to say 'large and well-
connected pore space'. This distinction becomes
important to the thesis of the paper, which is
that dolomitization does not create porosity.
However, it will be demonstrated that dolomi-
tization can create pore space and modify pore
size.
The problem
The principal culprit responsible for the
confusion is the mole-for-mole replacement
equation for dolomitization:
Mg +2 + 2CaCO 3 = MgCa(CO3) 2 + Ca +2. (1)
Using this equation, De B e a u m o n t (1837)
demonstrated that the mole-for-mole
conversion of calcite to dolomite results in a
12% decrease in volume because the molar
volume of dolomite is smaller than the molar
volume of calcite. He proposed, therefore, that
epigenetic dolostones should have 12%
porosity. He argued that this fact explains the
presence of vugs observed in the dolostones in
the Tyrol. The equation has been the model for
dolomitization most used since that time.
The problem with this model is that it
assumes that only magnesium, and no calcium
or carbonate, is added from the dolomitizing
water. Weyl (1960), using a conservation of mass
argument, attempted to demonstrate that no
carbonate could be added from the formation
waters because of the small volume of carbon-
ate found in groundwaters. However, it has
recently been demonstrated that calcium and
carbonate are extracted from the dolomitizing
fluid, along with magnesium (Lucia & Major
1994). In addition, there are numerous obser-
vations of 'pore-filling dolomite' in limestones
and dolostones. Gregg et al. (1993) demon-
strated that a large volume of pore space was
occluded by late dolomite. This pore-filling
dolomite did not require the replacement of
CaCO3 to precipitate. Murray (1960) addressed
this problem by introducing the concept of local
and distant sources of carbonate, local source
being equivalent to the volume of carbonate
available from the dissolution of the local lime-
stone and distant source being carbonate trans-
ported in with the dolomitizing fluid. Although
Murray (1960) envisioned that distant-source
dolomitization was a later phase, it is more
reasonable to assume that it would be active to
some degree during the replacement phase.
Halley & Schmoker (1983) demonstrated that
young limestones in south Florida are very
porous and suggested that the associated dolo-
stones inherited their porosity from the precur-
sor limestone. They proposed a pore-filling
phase to account for young dolostones having
less porosity than associated limestones and
introduced the term overdolomitization to
describe the pore-filling phase. A more correct
chemical model for dolomitization is therefore
one that includes the addition of carbonate
and magnesium during the replacement phase,
and the addition of calcium, magnesium and
carbonate during the pore-filling phase.
Morrow (1990) proposed the following
equation (equation 2) for volume-for-volume
replacement. The volume of carbonate needed
to replace the difference in molar volumes is 0.1
for aragonite and 0.25 for calcite. It is reason-
able to assume that this small amount of carbon-
ate will continue to be available for dolomite
precipitation after all the calcite or aragonite
has been replaced until the water reaches equi-
librium with dolomite and calcite/aragonite.
This is the pore-filling phase, also referred to as
'overdolomitization' by Halley & Schmoker
(1983) and Lucia & Major (1994) (equation 3).
Replacement phase:
(2 - x)CaCO3 + Mg 2+ + x C O 3 -2 =
CaMg(CO3)2 + (1 - x ) C a 2+ (2)
where x -= moles of carbonate added from the
dolomitizing fluid.
Pore-filling phase (overdolomitization):
Ca 2+ + Mg 2+ + 2CO3 = CaMg(CO3)> (3)
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
DOLOSTONE PORE SPACE
Approach
The thesis of this paper is that dolostone
porosity is related to the porosity of the precur-
sor limestone and the amount of carbonate
added from the dolomitizing fluid. The
approach taken to investigate this thesis focuses
on three basic questions: (1) What was the
porosity of the precursor limestone? (2) What
volume of carbonate has been added during and
after the replacement phase? (3) How much
porosity has been lost to compaction? By
limiting the investigation to young carbonates
from non-tectonic settings, the effects of
compaction and late dolomitization are mini-
mized, allowing work to focus on the nature of
the precursor limestone and the volume of
carbonate added.
Porosity o f p r e c u r s o r limestones
The porosity of dolomitized limestone will
depend on whether dolomitization occurred
early or late in the diagenetic history of a lime-
stone. It can occur syndepositionally or it can
occur 160 Ma after deposition, as described by
Kupecz & Land (1991) for the Ellenburger
(Lower Ordovician) of west Texas. It is well
known that carbonate sediments are very
porous, having porosity values that range from
45 to 70%, depending on the volume of lime
mud (Enos & Sawatsky 1981). Limestones do
not have porosity values higher than this range
and, with rare exception, have porosities much
lower. Schmoker et al. (1985) reported an
average porosity value of 12% for limestone
reservoirs in the United States with a range of
9-17%. The basic question, therefore, is not
how limestone porosity was created but how the
porosity was destroyed and at what rate. The
data presented here are from passive margins
because the porosity history of tectonically
active areas is complicated by the high compres-
sive forces and temperatures typical of that
environment. For example, Miller & Folk
(1994) reported a porosity of 0.1 percent for the
thrust and folded Triassic limestones of
northern Italy.
In passive margins limestones tend to retain
their porosity, losing it only gradually over
geological time by cementation and
compaction. Grainstones of Eocene age in
south Florida located between depths of 92 and
490 m have an average porosity of 32%, and
wackestone 36% (Budd 2002) (Fig. 1). Grain-
dominated fabrics and wackestones from
Pliocene-Miocene limestones at depths mostly
between 200 and 600 m on the Great Bahama
143
Bank both have an average porosity of 40%
(Melim et al. 2001) (Fig. 1). Neither of these
examples have evidence of meteoric diagenesis.
In contrast, outcrops of Plio-Pleistocene lime-
stone on Bonaire, Netherlands Antilles, have an
average porosity of 25% (Lucia & Major 1994).
Here there is ample evidence of extensive
meteoric cementation and dissolution that
resulted in the low porosity values.
Porosity in Cretaceous and Jurassic lime-
stones commonly decreases as mud content
increases, a phenomenon illustrated by Loucks
(2002) using data from Cretaceous limestones in
south Texas at depths ranging from 2000 to
2350m. Grain-dominated fabrics have an
average porosity of 13%, whereas mud-
dominated fabrics average 7% porosity (Fig. 2).
Cruz (1997) described Cretaceous reservoirs
found at a depth of 4750 m in the Santos Basin,
offshore Brazil, that have grain-dominated
fabrics with 16% average porosity and mud-
dominated fabrics with 2% average porosity.
The same cannot be said for the Cretaceous
Shuaiba mud-dominated fabrics in the Middle
East that typically have high porosity values. For
example, in the Yibal Field, Oman, mudstones
and wackestones have an average porosity of
35% at a depth of around 1450 m. In contrast,
Rudist grain-dominated fabrics found in the A1
Huwesiah Field have less porosity, averaging
18% (Fig. 2). High porosity in the mud-
dominated fabrics is not unusual when compared
with that of the Tertiary of Florida and the Great
Bahama Bank, but it is unusual when compared
with other Cretaceous carbonates. The Yibal
Field is the shallowest of the three Cretaceous
examples presented here, and the high porosity
most probably results from a lack of deep burial
or lack of cement, as suggested by Moshier
(1989). Other investigators have suggested that
the high porosity of the Shuaiba mud-dominated
fabrics results from dissolution (Harris & Frost
1984). My observations, however, agree with
those presented by Moshier (1989), who
concluded that the high porosity did not result
from dissolution but from incomplete cementa-
tion. The lower porosity value in rudist, grain-
dominated fabrics most probably results from
meteoric diagenesis of the rudist shoal.
An example from the Jurassic Arab D reser-
voir in the Ghawar Field, Haradh area, Saudi
Arabia, from depths between 2100 and 2200 m,
is illustrated in Figure 3. Although the data are
highly scattered, average values show a trend of
decreasing porosity with increasing lime mud
(Lucia et al. 2001), a situation similar to that
commonly observed in Cretaceous limestone
reservoirs.
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

144 F.J. LUCIA

Fig. 1. Porosity histograms for grain-dominated and mud-dominated Tertiary limestones. (a) Palaeogene of
west-central south Florida (Budd 2002). (b) Neogene of the Great Bahama Banks (Melim et al. 2001).

These data are consistent with the conclusion
of Schmoker & Hally (1982) that, in a passive-
margin environment, limestones lose their
porosity gradually, probably related to their
burial history (Fig. 4). Limestones less than
40 Ma old can have 30-40% porosity. The excep-
tion is limestone that has experienced extensive
meteoric diagenesis, in which case porosity is
reduced to the 20-30% range. There is no
distinction between grain- and mud-dominated
fabrics. Beginning with Cretaceous limestone,
however, mud-dominated fabrics gradually
become less porous than grain-dominated
fabrics. In the author's experience, this trend
continues into Palaeozoic rocks, where mud-
dominated fabrics are commonly dense and
grain-dominated fabrics are commonly porous.
Porosity o f associated dolostones
Limestones lose their porosity gradually over
tens of millions of years, and dolomitization can
occur syndepositionally or millions of years
after deposition. The porosity of the precursor
limestone will, therefore, partly depend on the
time of dolomitization. Determining the age of
dolomitization relative to the host limestone is
never easy. The dolomitization of peritidal sedi-
ments is assumed to be syndepositional.
Dolomitized subtidal facies that are continuous
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

DOLOSTONE PORE SPACE 145

Fig. 2. Porosity data for Cretaceous limestones. (a) Plot of porosity v. lime mud content for south Texas
Cretaceous samples, showing that average porosity decreases with increasing lime mud content (Loucks 2002).
(b) Porosity histograms for Shuaiba (Cretaceous) samples from Oman. Mud-dominated samples from theYibal
Field have an average porosity of 35%. Rudist grainstone samples from the A1 Huwiasiah Field have an
average porosity of 18%.

Fig. 3. Cross-plot from the Arab D reservoir in the Haradh area of the Ghawar Field, Saudi Arabia. (a)
Porosity v. rock fabric. GDP: grain-dominated packstone; MDP: mud-dominated packstone. (b) Porosity v. per
cent lime mud. Although there is considerable scatter in the data, the average trends show decreasing porosity
with increasing lime mud content (Lucia et al. 2001).

with and lie b e n e a t h peritida! d o l o s t o n e are to s o m e o t h e r datable d i a g e n e t i c or structural

generally t h o u g h t to be the same age as the
tidal-flat dolomite. T h e age of subtidal dolo-
s t o n e b o d i e s n o t c o n t i n u o u s with tidal-flat
dolostones is difficult to d e t e r m i n e , and the time
of d o l o m i t i z a t i o n is often d e t e r m i n e d b a s e d on
g e o c h e m i c a l data or relating d o l o s t o n e bodies
event. Oxygen and strontium isotopes have
b e e n used to give a P e r m i a n age to the dolomi-
tization of P e r m i a n c a r b o n a t e s in west Texas
(Bein & L a n d 1982). D o l o m i t i z a t i o n of the E1
P a s o G r o u p o c c u r r e d a f t e r c o l l a p s e of an
O r d o v i c i a n c a v e r n system, a c o l l a p s e t h a t
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
146
lOO
i
s
O
13_
50
0 I I I I I
50
Ume m u d (%)
Fig. 4. Plot of per cent lime mud v. per cent porosity for Holocene, Neogene, Palaeogene, Cretaceous and
Jurassic limestones. The plot shows: (1) the gradual loss of porosity with time (and presumably burial); and (2)
greater porosity loss with increasing mud content starting in the Cretaceous, or 100 Ma after deposition.
continued through the Silurian (Lucia 1995a).
Late dolomitization, therefore, occurred at least
100 Ma, and perhaps as much as 160 Ma, after
deposition, assuming the Pennsylvanian age for
the dolomitization proposed by Kupecz & Land
(1991) for the time-equivalent Ellenburger
Group. Wilson et al. (1990) dated late dolomiti-
zation in the Italian Alps by physical relation-
ships to igneous intrusives, and Zempolich &
Hardie (1997) dated late dolomitization by
physical relationships to Alpine folding and
thrusting.
To minimize the possible time gap between
deposition and dolomitization, porosity infor-
mation is collected from young carbonates
where limestones and dolostones are juxta-
posed. The youngest dolomite known is
Holocene in age and is commonly found near
the surface in evaporitic tidal-fiat deposits.
Unfortunately, there is little porosity infor-
mation published about Holocene dolomitic
sediments or dolostones. What few data are
available have been published by Lucia (1999)
and indicate that Holocene dolomitic sediments
are very porous (Fig. 5), showing no trend of
increasing or decreasing porosity through
dolomitization. Clearly, porosity is not created
by dolomitization but inherited from the
sediment. Lasemi et al. (1989) concluded, on the
F.J. LUCIA
Holocene
- - E n o s and Sawatsky (1981)
Miocene/Pliocene
~ , Melim et al. (2001)
Eocene/Oligocene
David Budd (2002)
Cretaceous Shuaiba
PDO data
.,,,,,,.Jurassic Arab D
~ ,f Lucia (2001)
I I I I Cretaceous Pearsall
100 Loucks (2002)
basis of textural evidence, that Holocene
dolomite from Andros Island is pore filling.
However, no porosity data were presented to
demonstrate porosity gain or loss.
The transition from dolostone to limestone
found in Plio-Pleistocene carbonate outcrops on
Bonaire, Netherlands Antilles, was described by
Lucia & Major (1994). They attributed dolomi-
tization to the evaporite reflux model and
concluded that these carbonates have never
been buried. The limestone is composed of
grain-dominated fabrics with an average
porosity of 25% and a porosity range of 10-40%
(Fig. 5). Porosity values are lower than those
found in carbonate sediments because of the
extensive meteoric diagenesis during Pleisto-
cene low stands (Lucia & Major 1994). The
porosity values of dolomitic limestones range
between 20 and 30%, a range similar to that of
limestone porosity. The adjacent dolostones
have an average porosity of 11% and a porosity
range of 3-30%. If magnesium only and no
carbonate were added during dolomitization, as
is assumed by the mole-for-mole replacement
model, the average dolostone porosity should
be 35% instead of 11% (Fig. 6). The mole-for-
mole dolomitization equation therefore does
not explain porosity values. Some carbonate, as
well as magnesium, must be added to keep the
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

DOLOSTONE PORE SPACE 147

Fig. 5. Ternary diagrams of porosity, dolomite and CaCO3 data from the Holocene and Plio-Pleistocene. (a)
Holocene data from Qatar and Bonaire showing no recognizable change in porosity with increasing amounts
of dolomite (Lucia 1999). (b) Plio-Pleistocene data from Bonaire showing no increase in porosity during
calcite replacement and loss of porosity during overdolomitization (Lucia & Major 1994).

Fig. 6. Porosity-frequency plots for dolostone, the precursor limestone and a hypothetical dolostone
calculated from the limestone frequency plot, assuming mole-for-mole replacement (Lucia & Major 1994),
showing that the mole-for-mole model does not explain the dolostone porosity.

porosity from increasing during the replace-
m e n t phase, and calcium, magnesium and
carbonate must be added after the replacement
phase to reduce average porosity to 11%.
Dolostone of Miocene age (McNeill et al.
2001) is found in the U n d a well drilled on the
Great Bahama Bank at a depth of 270-354 m.
The dolostone has an average porosity of 40%
(Melim et al. 2001), which is equal to the
average porosity of limestones in the U n d a and
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

148 F.J. LUCIA

Fig. 7. Porosity histograms for (a) Neogene limestones from the Unda and Clino wells and (b) Miocene
dolostones from the Unda well, Great Bahama Banks (Melim et al. 2001). Both limestone and dolostone have
40% porosity, showing that dolostone porosity is inherited from limestone.

Clino wells (Fig. 7). Little is understood about
the origin of the dolomitizing water, except that
it was most probably modified seawater, and
little is known about the time of dolomitization
(Swart & Melim 2000). There is no evidence
that these carbonates were ever subjected to
extensive meteoric diagenesis. The originally
high porosity of about 70% in the wackestones
was therefore reduced to the present value of
40% by simple burial c e m e n t a t i o n and
compaction, whereas the grain-dominated
fabrics retained their depositional porosity of
40%. It is likely that dolomitization occurred
after the wackestones were compacted to 40%
porosity. The high porosity in muddy sediments
is lost in the first 100 m of burial, according to
Goldhammer (1997), indicating that dolomiti-
zation occurred after 100 m of burial. The simi-
larity between limestone and dolostone porosity
values suggests that: (1) dolostone porosity was
not created through the process of dolomitiza-
tion but was simply inherited from the precur-
sor limestone: and (2) no carbonate was added
after the replacement phase of dolomitization.
Unfortunately, no limestone-dolostone pairs
are available from Cretaceous reservoirs. A
search of the literature indicates that there are
but a few Cretaceous dolostone reservoirs. Data
from the Arab D reservoir in the Ghawar Field,
Saudi Arabia, of Jurassic age provide insight
into the porosity history of interbedded lime-
stones and dolostones. Rock fabrics from the
Arab D reservoir were originally described by
Powers (1962) and more recently by Lucia et al.
(2001). There is no indication of extensive
meteoric diagenesis. The age of dolomitization
and the nature of the dolomitization water are
poorly understood, although geochemical data
suggest modified marine water (Cantrell et aL
2001). A plot of dolomite as a per cent o f b u l k
v o l u m e (BV) v. porosity shows that porosity
tends to decrease as per cent BV dolomite
increases (Fig. 8). At about 80% BV dolomite,
porosity increases to a point similar to that of
the porosity of the limestone. From this point
on, the samples are all dolostone, except for a
few anomalous calcitic dolostones, and the
porosity of the dolostone decreases. This
relationship is similar to one described by
Powers (1962). The decrease in porosity from 0
to 80% BV dolomite is attributed to an increase
in mud-dominated fabrics and a corresponding
increase in compaction. The compaction is
thought to occur much later than dolomitiza-
tion, and the porosity of the precursor limestone
and dolomitic-limestone fabrics is suggested to
be between 30 and 40%. Carbonates with more
than 80% BV dolomite resisted compaction
because the dolomite forms a supporting frame-
work. The grain-dominated fabrics also resisted
compaction. The similarity between the
porosity of grain-dominated limestones and
dolostones suggests that dolostone porosity is
inherited from the precursor limestone. The loss
of porosity in the dolostone is attributed to vari-
ations in the porosity of the precursor and to the
addition of carbonate after the replacement
phase (overdolomitization).
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

DOLOSTONE PORE SPACE 149

Fig. 8. A plot of porosity v. bulk-volume dolomite (dolomite as a per cent of mineral volume plus porosity) for
various fabrics, showing that average porosity decreases with increasing dolomite until little or no calcite
remains. This loss of porosity is attributed to differential compaction related to the increase in mud-dominated
fabrics, which are most susceptible to compaction. Dolostone porosities can be as high as limestone porosity
but decrease as dolomite cement is added (overdolomitization).

The data from H o l o c e n e and Tertiary lime-

Porosity stone-dolostone pairs demonstrate that young
100% dolostones inherit porosity from precursor lime-
stones (Fig. 9). The Miocene example has less
porosity than the Holocene example reflecting
d o l o m i t i z a t i o n after shallow burial. The
Holocene /
, ......... , \ / \ /L\ Bonaire example demonstrates that if the lime-
stone has u n d e r g o n e extensive meteoric dia-
Plio-Pleistocene )z._.._..~\ /
(Lucia and Major 1994) /
\\
Moc~n~
~ -- ~'
genesis, with accompanying porosity loss,
~ / ~ - - / \ .~,//~( Mebiiamnet
,ma2aoSo
a I) porosity in the dolostone reflects this loss. In
addition, dolomite cementation continues after
Jurassic~ ~ / ~ / "~ r e p l a c e m e n t dolomitization. The Jurassic
example illustrates the effect of burial diagene-
sis and overdolomitization on porosity values.
The dolostones resist compaction and preserve
porosity. However, the addition of pore-filling
100% 100% d o l o m i t e after r e p l a c e m e n t reduces porosity
CaCO3 Dolomite below the porosity of the precursor limestone.
The m u d - d o m i n a t e d limestones, and to some
Fig. 9. Ternary diagram of porosity, CaCO3, and d e g r e e g r a i n - d o m i n a t e d limestones, lose
dolomite from the Holocene, Plio-Pleistocene, porosity through compaction. Porosity reduc-
Miocene and Jurassic. There is little difference tion with increasing dolomite is therefore not
between limestone and dolostone porosity in the related to the dolomitization process but to the
Holocene and Miocene. The decrease in porosity
with increasing dolomite in the Jurassic is attributed susceptibility of m u d d y sediments to com-
to differential compaction of limestone v. dolostone. paction.
Both Plio-Pleistocene and Jurassic data show loss of A recent defence of the mole-for-mole m o d e l
porosity in the dolostone through was offered by Wilson et al. (1990) and
overdolomitization. Zempolich & Hardie (1997). In both papers the
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
150 F.J. LUCIA
precursor limestone is assumed to be dense, but
no porosity values are reported for the current
limestone and no discussion of the diagenetic
history of the limestone is presented. Wilson
et al. (1990) referred to 'porous dolomite' but
offered no porosity measurements to substanti-
ate or quantify this observation. Zempolich &
Hardie (1997) presented a well-documented
history of porosity evolution observed at lime-
stone-dolomite contacts. Visible porosity in thin
section and polished slabs was determined to be
10-15% using point-counting methods. Micro-
pores are not visible in thin sections of slabs, so
point-count porosity values are typically under-
stated. The large crystal and grain sizes of these
fabrics suggest, however, that the visible
porosity values are close to total porosity. The
age of the pore space relative to dolomitization
is not clear. Most of the pores are oomolds that
can be reasonably assumed to have formed
early in the diagenetic history of the limestone
and before dolomitization. It is nevertheless
argued that oomolds are found only in associ-
ation with the dolomitization front. Also, the
limestone is a deep-water oolite shed off a
shallow-water platform, minimizing the likeli-
hood of oomolds being formed by meteoric
dissolution. In addition, oomolds are commonly
produced by dissolution of unstable aragonite,
and aragonite is uncommon in the greenhouse
environment of the Triassic. Interestingly, both
these papers focus on diffusion of magnesium
for dolomitizing the dense limestone, as did
Miller & Folk (1994). This is a very different
dolomitization model from the fluid-flow model
envisioned in this paper and could result in no
additional carbonate being available during
dolomitization.
Petrophysical properties
If dolostone simply inherits porosity from the
precursor limestone, and may actually reduce
porosity through overdolomitization, why are
dolostones often better reservoirs than lime-
stones? One reason is that dolostones are less
susceptible to compaction than limestones,
maintaining their porosity more effectively.
Another reason is the change in pore structure
that occurs during dolomitization. Permeability
and water saturation are two important petro-
physical parameters that describe a carbonate
reservoir. They are controlled fundamentally by
pore-size distribution, and pore-size distribution
has been shown to be controlled by particle size
and sorting (Lucia 1995b). The pore-size distri-
bution in a dolograinstone is similar to that in
the precursor limestone and typically has little
relationship to crystal size (Lucia 1995b). The
pore space and porosity are inherited from the
precursor limestone, not created through
dolomitization. Similarly, the pore-size distri-
bution in grain-dominated dolopackstones is
controlled by the grain size and sorting of the
precursor limestone, and the pore space and
porosity are inherited from the precursor lime-
stone.
In m u d - d o m i n a t e d dolostones, however,
crystal size pays a dominant role in pore-size
distribution (Fig. 10). As the dolomite crystal
size increases, the pore size also increases. In
this case, the porosity of the dolostone is inher-
ited from the precursor limestone, but the pore
space is newly created. If a wackestone is
converted to dolostone with a fine (less that
20 ~m) crystal size, the dolostone will have pore
sizes similar to those of the precursor limestone
because lime mud is commonly composed of
crystals and grains that are less than 20 ~m in
size. However, if the dolomite crystals are larger
than 20 ~m, the pore size will be larger than that
of the precursor limestone, and permeability
will be higher and water saturation lower for
equivalent conditions. Assuming that inter-
particle porosity is constant, permeability of
mud-dominated dolostones is a direct function
of dolomite crystal size, and permeability
increases by about a factor of 10 as crystal size
increases from fine to medium to large (Fig. 10).
Figure 11 illustrates the process of converting
a wackestone with micropores to a medium
crystalline dolowackestone with large intercrys-
tal pores. As dolomite is formed, it replaces not
only the lime mud particles but also pore space
between them. The adjacent lime mud is
dissolved to supply most of the carbonate
needed to fill this pore space. A small amount
of carbonate comes from the dolomitizing water
as well, keeping the porosity constant. As
dolomitization progresses, the crystals impinge
on one another, and intercrystal pore space is
created. The size of this pore space is controlled
by the size of the dolomite crystals and is larger
than the micropores. When all or most of the
lime mud has been dissolved, carbonate must be
supplied to the growing rhombs by dissolution
of the large allochems, resulting in the
formation of grain moulds, a type of separate
vug (Lucia 1995b, 1999). Porosity is not
increased by this process because the dissolved
carbonate is reprecipitated locally on the
growing dolomite crystals, along with a small
amount of carbonate from the water. At this
stage, the wackestone and dolowackestone have
the same porosity but different pore space; the
pore space in the dolowackestone was created
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

DOLOSTONE PORE SPACE 151

10

[] Xl size <20 pm Class 1

9 XI size 20-1 00 lam ~ 4- 4-+/,,"
4 Xl size >100 tam / -r_t/ T. /

4- / 4- 4-4- ~- ~ / Class 2

Cl
E
4- / 4-
, ~4-
4- 9
v ~ 4-44_4- ,-t- , ~ : - I v '" //u Class 3
..i-.,
. m

..ID

E
10 -h
13_ 4- -t-/ 7- ~. / 9 / 9 ,,u[] L_~ /"

-t- /4- 4- .. oo~ [] []

4-
4- / 4/- 9 4- 9 q - o
~ /"-o 9 ~ /v'eoo
~~ "_ ~176
-s ~ []
1_ ~0.,'_o-o- [] / " Do

4- -I-
-h~- o+
0.1- / 9 i i I
6 8 0 30
Interparticle porosity (%)

Fig. 10. Cross-plot of interparticle porosity (intercrystal porosity) and permeability showing the effect of
crystal size in mud-dominated dolostones (Lucia 1995b). Petrophysical classes 1, 2 and 3 are shown for
comparison purposes.

through dolomitization. After all the limestone
has been dissolved, the d o l o m i t e crystals
continue to grow using magnesium, calcium and
carbonate from the dolomitizing water until the
water reaches equilibrium. This overdolomi-
tization can reduce porosity to 5 % or less. With
burial, the partially dolomitized wackestones
lose porosity by compaction, whereas the dolo-
stones resist compaction and retain their
porosity.
Figure 12 is a cross-plot of interparticle
porosity and permeability for all dolostones.
Large crystalline dolostones with any precursor
fabric, as well as dolograinstones, plot in the
petrophysical class 1 field (see Lucia 1995b).
Medium crystalline, mud-dominated dolostones
and grain-dominated dolopackstones plot in the
petrophysical class 2 field, and fine crystalline,
mud-dominated dolostones plot in the petro-
physical class 3 field. According to Lucia
(1995b), grainstone fabrics plot in the class 1
field, grain-dominated packstones plot in the
class 2 field and mud-dominated fabrics plot in
the class 3 field. The effect of dolomitization on
petrophysical properties is to increase the petro-
physical characteristics of m u d - d o m i n a t e d
fabrics from class 3 to classes 2 and 1. In a dolo-
stone reservoir the mud-dominated fabrics are
therefore commonly excellent reservoir rocks,
along with grain-dominated dolostones,
whereas in a limestone reservoir only grain-
dominated fabrics have excellent reservoir
quality.
Conclusions
The data presented here suggest the following
model for porosity development in dolostones
in a passive-margin environment. Assuming
that dolomitization occurred within the first
40 Ma after deposition, it can be reasonably
assumed that the limestone porosity was in the
30-40% range. During the replacement phase,
the dolostone will inherit this porosity and any
change in molar volume will be offset by the
precipitation of pore-filling dolomite using
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

152 F.J. LUCIA

0 0 0

,Z::l ' - - 9 0

~.~ ~

~ ~.~

~ .~-~

~'~

' ~ o ~
ol/h

9 9
-6--;A

9~ - s ~ 2 ~-~

9-" o-a~ ~'a

~ o o t ~ ~'o-a

N 2 u ~
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015

DOLOSTONE PORE SPACE 153

Fig. 12. Cross-plot of interparticle porosity (intercrystal and intergrain porosity) and permeability comparing
distribution of the five basic dolostone rock fabrics with the petrophysical class fields. Dolograinstone and
large crystalline (Lxl) dolostone plot in the class I field, grain-dominated dolopackstone and medium
crystalline mud-dominated dolostone (Mxl mdf) plot in the class 2 field, and fine crystalline mud-dominated
dolostone (Fxl mdf) plot in the class 3 field (Lucia 1995b).

carbonate t r a n s p o r t e d in the dolomitizing
water. After the replacement phase, dolomite
continues to precipitate in the pore space and
overdolomitization occurs reducing porosity to
as low as 5%.
The dolomitization model must account for
the addition of magnesium as well as carbonate
during the replacement and the overdolomi-
tization phase.
Replacement phase (Morrow 1990)
(2 - x)CaCO3 + Mg 2+ + xCO32- =
CaMg(CO3)2 + (1 - x)Ca 2+
where x = moles of carbonate added from the
dolomitizing fluid.
Pore-filling phase (overdolomitization)
Ca 2+ + Mg 2+ + 2CO3 = CaMg(CO3)2.
Limestone porosity is lost to compaction and
cementation very gradually over tens of millions
of years, with mud-dominated fabrics losing
porosity faster than grain-dominated fabrics.
Dolostones resist compaction more than lime-
stones, although initially dolostones will have
the same or less porosity than adjacent lime-
stones. The limestone will lose porosity prefer-
entially with burial owing to compaction, result-
ing eventually in porous dolostones and dense
limestones. This process, however, will take
h u n d r e d s of millions of years in a passive-
margin setting.
This model is based on an open-circulation
system in which flowing dolomitizing water
brings in magnesium, calcium and carbonate
hydrodynamically. In a diffusion-dominant
dolomitization model, it is possible that only
magnesium and calcium are transported in and
(2) out of the limestone along the dolomitization
front and that carbonate is not available to form
additional dolomite to c o m p e n s a t e for the
molar volume changes. As long as there is fluid
flow, however, the availability of carbonate
from the water to add dolomite to offset the
(3) volume change must be considered.
Dolostones commonly make excellent reser-
voirs because they retain porosity and because
m u d - d o m i n a t e d dolostones can have larger
pore sizes than m u d - d o m i n a t e d limestones.
Grain-dominated dolostones tend to have the
same pore structure as the precursor limestone.
Dolomitized mud-dominated fabrics, however,
can have significantly larger pores than the
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
154 F.J. LUCIA
p r e c u r s o r limestone b e c a u s e d o l o m i t e crystal
sizes are c o m m o n l y larger t h a n the p r e c u r s o r
m u d size. P e r m e a b i l i t y can be p r e d i c t e d if inter-
particle (grains or crystals) porosity, the precur-
sor rock fabric and d o l o m i t e crystal size are
known.
I would like to acknowledge the many thoughtful
discussions with my colleagues C. Kerans, S. Ruppel
and R. Loucks here at the Bureau of Economic
Geology. The manuscript was edited by L. Dietrich to
whom I am deeply indebted. Financial support came
from the many sponsors of the Reservoir Charac-
terization Research Laboratory project, an Industrial
Associates project at the Bureau of Economic
Geology, John A. and Katherine G. Jackson School of
Geosciences, The University of Texas at Austin.
References
BEIN, A. & LAND, L.S. 1982. San Andres Carbonates
in the Texas Panhandle: Sedimentation and
Diagenesis Associated with Magnesium-
Calcium-Chloride Brines. The University of
Texas at Austin, Bureau of Economic Geology
Report of Investigations, 121.
BUDD, D.A. 2002. The relative roles of compaction
and early cementation in the destruction of
permeability in carbonate grainstones: a case
study of the Paleogene of west-central Florida,
U.S.A. Journal of Sedimentary Research, 72,
116-128.
CANTRELL, D.L., SWART P.K., HANDFORD, R.C.,
KENDALL, C.G. & WESTPHAL, H. 2001. Geology
and production significance of dolomite, Arab-D
reservoir, Ghawar field, Saudi Arabia. GeoArabia,
6, 45-60.
CRUZ, W.M. 1997. Study of Albian carbonate analogs:
Cedar Park Quarry, Texas, USA, and Santo Basin
reservoirs, southeast offshore Brazil. PhD disser-
tation, The University of Texas at Austin, Austin,
TX.
DE BEAUMONT, E. 1837. Application du calcul a
t'hypoth~se de la formation par epigenie des
anhydrites, des gypses et des dolomies. Bulletin
de la SociOtO GOologique de France, 8, 174-177.
ENOS, P. ~r SAWATSKY,L.H. 1981. Pore networks in
Holocene carbonate sediments. Journal of Sedi-
mentary Petrology, 51, 961-985.
GOLDHAMMER, R.K. 1997. Compaction and decom-
paction algorithms for sedimentary carbonates.
Journal of Sedimentary Research, 67, 26-56.
GREGG, J.M., LAUDON,P.R., WOODY, R.E. & SHELTON,
K.L. 1993. Porosity evolution of the Cambrian
Bonneterre Dolomite, Southeastern Missouri,
USA. Sedimentology, 40, 1153-1169.
HALLEY, R.B. & SCHMOKER,J.W. 1983. High-porosity
Cenozoic carbonate rocks of south Florida:
progressive loss of porosity with depth. AAPG
Bulletin, 67, 191-200.
HARRIS, P.M. & FROST, S.H. 1984. Middle Cretaceous
carbonate reservoir, Fahud Field and Northwest-
ern Oman. AAPG Bulletin, 68, 649-658.
KUPECZ, J.A. & LAND, L.S. 1991. Late-stage dolomi-
tization of the Lower Ordovician Ellenburger
Group, West Texas. Journal of Sedimentary
Petrology, 61, 551-574.
LANDES, K.K. 1946. Porosity trough dolomitization.
AAPG Bulletin, 30, 305-318.
LASEMI, Z., BOARDMAN, M.R. & SANDBERG, P.A.
1989. Cement origin of supratidal dolomite,
Andros Island, Bahamas. Journal of Sedimentary
Petrology, 59, 249-257.
LOUCKS, R.G. 2002. Controls on reservoir quality in
platform-interior limestone through the Gulf of
Mexico: example form the Lower Cretaceous
Pearsall Formation in South Texas. Gulf Coast
Association of Geologic Societies Transactions,
52, 659-672.
LUCIA, F.J. 1995a. Lower Paleozoic development,
collapse, and dolomitization, Franklin Mountains,
E1 Paso, Texas. In: BUDD, D.A., SALLER,A.H. and
HARRIS, P.A. (eds) Unconformities and Porosity in
Carbonate Strata. American Association of Petrol-
eum Geologists, Memoir, 63, 279-300.
LUCIA, F.J. 1995b. Rock-fabric/petrophysical classifi-
cation of carbonate pore space for reservoir
characterization. AAPG Bulletin, 79, 1275-1300.
LUCIA, F.J. 1999. Carbonate Reservoir Characteriz-
ation. Springer, Berlin.
LuciA, F.J. & MAJOR, R.P. 1994. Porosity evolution
through hypersaline reflux dolomitization. In:
PURSER, B.H., TUCKER, M.E. & ZENGER, D.H.
(eds) Dolomites: A Volume in Honour of
Dolomieu. International Association of Sedi-
mentologists, Special Publications, 21, 325-341.
LUCIA, F.J., JENNINGS,J.W., JR, RAHNIS,M. & MEYER,
F.O. 2001. Permeability and rock fabric from
wireline logs, Arab-D Reservoir, Ghawar Field,
Saudi Arabia. GeoArabia, 6, 619-646.
MCNEILL, D.F., EBERLI, G.P., LIDZ, B.H., SWART,P.K.
& KENTER,J.A.M. 2001. Chronostratigraphy of a
prograded carbonate platform margin: a record
of dynamic slope sedimentation, western Great
Bahama Bank. In: G|NSBURG, R. (ed.) Subsurface
Geology of a Prograding Carbonate Platform
Margin, Great Bahama Bank: Results of the
Bahamas Drilling Project. Society of Economic
Paleontologists and Mineralogists, Special Publi-
cations, 70, 61-100.
MELIM, A.L., ANSEI.METTI,R.S. & EBERLI, G.P. 2001.
The importance of pore type on permeability of
Neogene carbonates, Great Bahama Bank. In:
GINSBUR(~, R. (ed.) Subsurface Geology of a
Prograding Carbonate Platform Margin, Great
Bahama Bank: Results of" the Bahamas Drilling
Project. Society of Economic Paleontologists and
Mineralogists, Special Publications, 70, 217-240.
MILLER, J.K. & FOLK, R.L. 1994. Petrographic,
geochemical and structural constraints on the
timing and distribution of postlithification
dolomite in the Rhaetian Portoro ('Calcare
Nero') of the Portovenere Area, La Spezia, Italy.
In: PURSER, B.H., TUCKER, M.E. & ZENGER,
D.H. (eds) Dolomites: A Volume in Honour of
Dolomieu. International Association of Sedi-
mentologists, Special Publications, 21, 187-202.
 Downloaded from http://sp.lyellcollection.org/ at Carleton University Library on June 18, 2015
DOLOSTONE PORE SPACE
MORROW, D.W. 1990. Dolomite - part 1: The chem-
istry of dolomitization and dolomite precipi-
tation. In: MCILREATH, I.A. & MORROW, D.W.
(eds) Diagenesis. Geoscience Canada, Reprint
Series 4, 113-123.
MOSHIER, S.O. 1989. Development of microporosity
in a micritic limestone reservoir, Lower
Cretaceous, Middle East. Sedimentary Geology,
63, 217-240.
MU~tRAY, R.C. 1960. Origin of porosity in carbonate
rocks. Journal of Sedimentary Petrology, 30,
59-84.
POWERS, R.W. 1962. Arabian Upper Jurassic carbon-
ate reservoir rocks. In: HAM, W.E. (ed.) Classifi-
cation of Carbonate Rocks: A Symposium.
American Association of Petroleum Geologists,
Memoir, 1, 122-192.
SCHMOKER, J.W. & HALLEY, R.B. 1982. Carbonate
porosity versus depth: a predictable relation for
South Florida. AAPG Bulletin, 66, 2561-2570.
SCHMOKER, J.W., KRYSTINIK, K.B. & HALLEY, R.B.
1985. Selected characteristics of limestone and
dolomite reservoirs in the United States. AAPG
Bulletin, 69, 733-741.
155
SWART, P.K. & MELIM, L.A. 2000. The origin of
dolomites in Tertiary sediments from the margin
of Great Bahama Bank. Journal of Sedimentary
Petrology, 70, 738-748.
TUCKER, M.E. 2001. Sedimentary Petrology; An intro-
duction to the Origin of Sedimentary Rocks.
Blackwell Scientific, Oxford.
WEYL, P.K. 1960. Porosity through dolomitization:
conservation-of-mass requirements. Journal of
Sedimentary Petrology, 30, 59-84.
WILSON, E.N., HARDIE, L.A. & PHILLIPS, O.M. 1990.
Dolomitization front geometry, fluid flow
patterns, and the origin of massive dolomite: the
Triassic Latemar buildup, Northern Italy.
American Journal of Science, 290, 741-796.
ZEMPOLICH, W.G. & HARDIE, L.A. 1997. Geometry of
dolomite bodies within deep-water resedimented
oolite of Middle Jurassic Vajont limestone,
Venetian Alps, Italy: analogs for hydrocarbon
reservoirs created through burial dolomitization.
In: KUPECZ,J.A., GLUYAS,J.G. & BOLCH, S. (eds)
Reservoir Quality Prediction in Sandstones and
Carbonates. American Association of Petroleum
Geologists Memoirs, 69, 127-162.